CritTrans.Exe for the measurement and visualization of
critical transitions in the HistCite™ output
This routine computes the relative entropies generated by the transition between a set of matrices according to the Kullback-Leibler divergence (Kullback & Leibler, 1951; Theil 1972; Leydesdorff 1995). The results are combined with the output of HistCite™ to value the links in terms of probabilistic entropies.
(See Lucio-Arias, D., & Leydesdorff, L. (2008). Main-path analysis and path-dependent transitions in HistCite (TM)-based historiograms. [Article]. Journal of the American Society for Information Science and Technology, 59(12), 1948-1962. doi: 10.1002/asi.20903.)
The data matrices (similar and all fields numeric) can be made in Excel or SPSS, but they should be stored as dBASE-files. They should be numbered sequentially like MATR1.dbf, MATR2.dbf, etc. The maximum number of matrices is 1024, and there is a maximum of 1024 column vectors (e.g., words) per matrix. However, the number of rows (e.g., cited references) is unlimited. The program uses also the output file of HistCite™ in Pajek format. This file has to be renamed histcite.net and should be located in the same folder as the program and the input matrices.
If there are zeros in the data matrices, the user is advised to augment all values with one because otherwise a division by zero can generate an error message. (See Lucio & Leydesdorff for more details.)
One can choose between “(C)odification” or “(D)iffusion” given the directionality of the transitions which one wishes to analyze. (See also: Koen Frenken and Loet Leydesdorff (2000). Scaling Trajectories in Civil Aircraft (1913-1970), Research Policy 29(3), 331-348). The routine generates iij.dbf which contains the probabilistic entropies between each two matrices, hist2.txt which adds the expected information values (I) to the arcs in the HistCite file. This file can be visualized using Pajek.
Crtrans1.txt contains all possible transitions in the Pajek format, but the critical ones have a negative sign and are therefore dashed in the visualization using the default values in Pajek. Crtrans2.txt contains the valued links as arcs. One can color the negative links red using advanced output styles (e.g., svg).
The output files are overwritten in each round! Save them separately if useful.
@ Loet Leydesdorff,
Amsterdam, 10 Oct. 2019 (revised)